Predictive safety models for more autonomous systems in ATM

Due to the high level of safety achieved in ATM, relatively little data is available on negative safety outcomes. For this reason, the current knowledge of the determinants of the occurrence of Loss of Separation (LoS) incidents between aircraft is limited, as is the industry’s current ability to predict them.Thus, the aim is to develop a predictive methodology with the objective of characterising safety in airspace in terms of Separation Minima Infrigements (SMI) between en-route aircraft, based on the definition of models called Safety Performance Functions (SPF). For this purpose, Bayesian Inference has been selected for its high predictive capacity to estimate low probability events and for allowing the integration of knowledge modelling with data inference.A conceptual framework has been established to integrate the current available knowledge on causality and SMI precursors with the hindsight derived from the analysis of the type of data available.Finally, a sensitivity analysis is performed on the model with the intention of drawing conclusions on the most influential variables on the possible breach of ATM safety barriers that could lead to the occurrence of an SMI.